INT20CN Computer Networks

Tutorial #7

1. In the lecture, the statement was made that the various versions of HTTP are backwards compatible. What does this mean?

2. (Slightly tricky question) In HTTP/0.9, there was no way to specify the content type in objects returned from the server. How did the browser know whether it was receiving a Web page (in HTML), a GIF image, a sound sample or whatever?

3. In HTTP/1.0 the GET request (and other types as well, but we didn't elaborate in the lecture) is terminated by two newlines. Why are two necessary? Wouldn't one newline be sufficient, as in HTTP 0.9?

4. The response from HTTP/1.0 and HTTP1.1 servers is "MIME-compatible".
   1. What does this mean?
   2. What is the MIME type for ordinary Web pages in HTTP/1.0?
   3. Why is the "Content-length: " header required in HTTP/1.0?
   4. Why isn't an SMTP-style MIME "Content-encoding: " header required in HTTP/1.0 as it is for email?

5. A browser makes the following request to a Web server:

   GET /Fig1.gif HTTP/1.0<newline><newline>
   

   What would the server return? Explain in some detail.

6. What is the HTTP/1.0 HEAD request method used for?

7. A client can optionally include a GET request method header of the form If-Modified-Since:
   1. Why is this used, and what is it called?
   2. Why is the date/time specified in GMT (UTC) instead of local time?
   3. This header takes a date/time value. Is there an Internet standard format for this value? Hint: look at section 3.3 of RFC1945
   4. The HTTP/1.1 Etag provides a better mechanism for achieving the same result as an "If-Modified-Since:" request. Explain, very briefly how this works.

8. The HTTP/1.0 specification permits a GET request method to include a "Referer: " header in the request. Why is this considered to be a potential privacy issue?

9. Give some of the reasons why HTTP/1.0 is not a highly regarded protocol in the Internet technical community. How does HTTP/1.1 address these problems? At least two points required.

10. Proxy servers can reduce network (download) costs for an organisation by caching recently requested Web documents (pages, images, etc), however they are not altogether successful in practice. Give two reasons why proxy caches are not able to satisfy the majority of Web requests in The Real World(tm).

11. Contemplate this: the specification for the original version of HTTP (0.9) was approximately 6 Kbytes in size. The RFC for HTTP/1.0 (RFC1945) was 134 Kbytes. The RFC for the current version of HTTP/1.1 (RFC2616) is 412 Kbytes. What conclusions can you draw?